Skill-Based Treatment Cut Problem Behavior. 80% of Its Studies Rated Below Adequate.

Skill-based treatment (SBT) pairs functional communication training with delay and denial tolerance, usually built from an IISCA. A preregistered systematic review and meta-analysis in the Journal of Autism and Developmental Disorders, published September 28, pooled the single-case literature. The authors are Bunyan, Feuerer and Vostanis. They conclude SBT is “an emerging evidence-based practice.” They also say the results “suggest caution in considering SBT as evidence-based at this time.”

The finding

• Sample: 30 studies, 111 SBT applications, 107 participants. The abstract says 28 studies and 99 applications; the body gives the larger figures.
• Effect: baseline-corrected Tau averaged 0.47 (range −0.35 to 1.00). The abstract calls that “good.” By the authors' own bands, 0.20–0.60 is a moderate change.
• Spread: Tau-BC was computable for 59 participants. It was moderate for 25, large for 15, very large for 8 and small for 11. Forty-two had wide confidence intervals.
• Quality: 7% of studies rated borderline-strong, 13% adequate, 43% borderline-adequate and 37% weak. Among applications with narrow intervals, only three came from studies rated adequate or better.
• Dose: most applications ran under 20 hours a week. Where reported, total treatment averaged 68.78 hours and cost averaged $5,984.91.
• Social validity: questionnaires covered 73% of participants and were positive wherever asked.

The turn: who it's been tested on

96% of participants were under 14, and 80% were autistic. Most applications ran in outpatient clinics, delivered by therapists, graduate students or BCBAs. 95% used IISCA-derived contingencies. That's a narrow base for a procedure now used in homes and schools, with older clients, by RBTs.

The authors argue the other side too. They say the quality tool they used “can be overly sensitive,” scoring a study down for one participant's unstable data, which can mean a stronger design rather than a weaker one.

The limits, in the authors' terms

• Only one grey-literature database could be searched.
• Publication bias wasn't formally assessed.
• Tau-BC measures non-overlap, not magnitude.
• The second author delivers SBT training courses, a declared conflict of interest.

What you must know or do

• BCBAs writing plans that justify SBT as evidence-based: this is now the most current synthesis, and it says “emerging.” Check that your plan's language matches what it supports.
• For clients 14 or older, or SBT delivered at home by RBTs, you're outside most of the evidence base. Set a shorter data-review interval and spell out decision rules in the plan.
• Clinical directors: collect procedural fidelity on SBT cases. It's the first gap the authors ask the field to close.